© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 30040
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions or variables, particularly those involving negative values and powers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    177
    +0.14
    0.7%
    359
    +0.13
    0.7%
    172
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    506
    +0.14
    0.01
    316
    +0.13
    0.01
    248
    +0.11
    0.01
    Negative Logits
    ually
    -1.78
    äs
    -1.63
    ary
    -1.63
    »¿
    -1.61
    wald
    -1.57
    yrus
    -1.50
    igens
    -1.50
    ĻĤ
    -1.49
    unes
    -1.47
    opl
    -1.46
    POSITIVE LOGITS
    .)
    1.53
     Crime
    1.49
     Getting
    1.47
    ediatric
    1.44
     crew
    1.39
     hockey
    1.39
    .")
    1.37
    irds
    1.36
     liking
    1.34
     crime
    1.33
    Activations Density 0.380%

    No Known Activations